CherryStudio Setup Guide

A full-featured AI assistant desktop client.

Website: https://www.cherry-ai.com/ Download: https://www.cherry-ai.com/download

Overview

CherryStudio is a powerful AI assistant desktop app that supports many mainstream AI models, including Claude, Gemini, and GPT. It provides a unified AI interaction interface. This guide shows how to connect CherryStudio to Xiaomai API.

Features

CherryStudio provides:

• Multi-model support: Claude, Gemini, GPT, and more
• Unified interface: manage all AI services in one app
• Custom API: connect custom API providers
• Cross-platform: Windows, macOS, and Linux
• Local-first: data is stored locally for better privacy
• Rich features: conversation management, model switching, parameter tuning, and more

Installation

Step 1: Download and Install CherryStudio

1. Open the CherryStudio download page
2. Download the package for your system:
   • Windows: .exe installer
   • macOS: .dmg image
   • Linux: .AppImage or .deb
3. Install it following your system prompts

macOS Note

If macOS says the app cannot be opened because it is from an unidentified developer, allow it in System Settings -> Security & Privacy.

Step 2: Get a Xiaomai API Key

Before configuring CherryStudio, get an API key:

1. Open the Xiaomai API console
2. Sign in
3. Choose the auto group
4. Save the generated API key

Configure Provider

Step 1: Open Settings

1. Open CherryStudio
2. Click Settings or Preferences in the upper-right corner
3. Choose Model Services

Step 2: Add Provider

Click Add.

Configuration:

• Provider name: enter Xiaomai, or another name
• Provider type: choose OpenAI
• API Key: paste the key from the Xiaomai API console
• API address: https://xiaomai.win/v1/messages

Step 3: Add Models

Click Manage and select the models you want to use.

Start Using

Create a New Chat

1. Click New Chat
2. Choose a configured model from the model selector
3. Start chatting with AI

Switch Models

During a conversation:

1. Click the model selector at the top
2. Choose another configured model
3. Continue the conversation. Context may be preserved or reset depending on app settings

Tune Parameters

CherryStudio commonly supports:

• Temperature: controls randomness, usually 0 to 1
• Max Tokens: controls response length
• Top P: controls sampling range

Parameter Suggestions

• Coding tasks: Temperature 0.2-0.5 for better accuracy
• Creative writing: Temperature 0.7-0.9 for more creativity
• Daily chat: Temperature 0.5-0.7 for balance

Best Practices

1. Choose Models by Task

• Code writing: Claude Sonnet 4.5, claude-sonnet-4-5-20250929
• Fast chat: Gemini 3 Flash Preview, gemini-3-flash-preview
• Complex reasoning: Claude Opus 4.5, claude-opus-4-5-20251101
• Multimodal tasks: Gemini 3 Pro Preview with image support

2. Manage API Usage

• Check the balance in the Xiaomai API console regularly
• Create separate API keys for different use cases
• Do not expose API keys publicly

3. Improve Conversation Quality

• Use clear prompts
• Set context length reasonably
• Use conversation history management

Compared with Other Clients

Feature	CherryStudio	Alma	Hapi
Interface	Desktop app	Desktop app	Web/PWA
Multi-model support	Yes	Yes	Yes
Code editing	Partial	Yes	Yes
Terminal integration	No	Yes	Yes
Remote access	No	No	Yes
Learning curve	Low	Medium	Medium

Recommendations:

• Pure chat: CherryStudio, simple and easy to use
• Coding development: Alma or Hapi, with stronger development features
• Remote control: Hapi

FAQ

API key invalid

Possible causes:

• API key was entered incorrectly
• Balance is insufficient

Solutions:

1. Check whether the API key was copied completely
2. Check account balance

Model list is empty

Possible causes:

• Base URL is wrong
• Network connection issue
• API key lacks permission

Solutions:

1. Check network connectivity
2. Create a new API key and confirm token-group permissions

Responses are slow

Possible causes:

• Network latency
• Large model selected
• Context is too long

Solutions:

1. Check network quality
2. Try a faster model, such as Gemini Flash
3. Clear or reduce conversation history

How do I use multiple models?

In CherryStudio:

1. Configure separate model providers
2. Choose the target model when creating a new chat
3. Create multiple chat windows if needed, each using a different model

More Questions

See:

• Xiaomai API console
• Xiaomai API FAQ
• Support
• CherryStudio official docs

Done

Configuration is complete. Remember:

• Different model families may require API keys from different token groups
• Anthropic provider API address: https://xiaomai.win/v1/messages
• Gemini provider API address: https://xiaomai.win/v1beta/models
• Choose models by task
• Check your Xiaomai API console balance regularly
• Keep your API key safe

---

Related guides:

• Alma client setup
• Hapi remote control
• Xiaomai API console